Digital signal processing method based on quantum compressed sensing

Abstract

The invention discloses a digital signal processing method based on quantum compressed sensing, and belongs to the technical field of digital signal processing. Aiming at the digital signal processing requirements of broadband radio-frequency signal measurement and reconstruction, data compression rate improvement and the like, the method comprises the four steps of initial quantum state preparation, quantum state control, quantum state detection and signal reconstruction. Wherein the preparation of the initial quantum state is completed by a 1550nm narrow linewidth laser and a PDH frequency locking module; the quantum state control is completed by an intensity modulator, an electro-optical crystal and a direct current bias controller; quantum state detection is completed by a variable optical attenuator, a single photon detector, a time interval analyzer and a GPS locking quartz crystal. Signal reconstruction is completed by a data processing unit, a compressed sensing system is constructed by using a quantum physical process, compression measurement of sparse signals in a transform domain is realized, the real-time measurement bandwidth reaches 20GHz, the data compression rate reaches 10 <-6 >, and remarkable quantum advantages are shown.

Description

Technical Field

[0001] The present invention belongs to the technical field of digital signal processing, and in particular relates to a digital signal processing method based on quantum compressed sensing. Background Art

[0002] With the development of economy and society, digital signal processing technology is also developing rapidly. Digital signal processing technology mainly combines computers, digital circuits, microprocessors, data storage devices and displays to form a complete electronic information system. More and more industries are beginning to use digital signal processing technology in the construction of electronic information engineering. In the field of shortwave communication, digital signal processing technology simulates and processes the front-end signal, digitally processes the intermediate frequency signal, and finally outputs the three main signals of audio signal, AGC control signal and digital quantized baseband signal; in the field of graphics and images, digital signal processing technology can compress images to study the surface of the moon, analyze meteorological cloud maps, and study the composition of cosmic stars. Equipment such as infrared and millimeter wave radars in automotive electronic systems require digital signal processing technology for data analysis.

[0003] Traditional digital signal processing technology uses Nyquist sampling theorem to sample, ensuring that there is no distortion when recovering analog signals from discrete signals. The sampling frequency of the signal must be at least twice the signal bandwidth. However, in the field of high-frequency signal processing, signal processing technology based on this theory needs to face great challenges in massive data processing and transmission. High-speed sampling will generate a large amount of redundant information, resulting in a great waste of storage and transportation resources. This requires a method to compress the data, then sample it, and collect a small amount of useful information to reconstruct the original signal.

[0004] Compressed sensing theory is another major breakthrough in the field of digital signal processing after the famous Nyquist sampling theorem. It can sample at a sampling rate far lower than that required by the Nyquist theorem and restore sparse signals in the transform domain through reconstruction algorithms. The key to applying compressed sensing theory is the hardware implementation of the measurement matrix, of which random matrices are the most commonly used. However, it is impractical to implement a completely random matrix in hardware for classical electronics systems. Therefore, the hardware implementation of compressed sensing theory often adopts the method of artificially constructing a random measurement matrix. Although the sampling rate is reduced, its measurable bandwidth is still limited by hardware parameters such as the random number generation rate and ADC time resolution. More importantly, this artificially constructed perception matrix is ​​independent of the measured signal, so classical compressed sensing is essentially a non-adaptive measurement. Summary of the invention

[0005] In response to the needs of processing digital signals such as broadband RF signal measurement and reconstruction and improving data compression rate, the present invention provides a digital signal processing method based on quantum compressed sensing.

[0006] In order to achieve the above object, the present invention adopts the following technical solutions:

[0007] A digital signal processing method based on quantum compressed sensing, the method comprising the following steps:

[0008] Step 1, initial quantum state preparation: the narrow linewidth laser outputs laser light into the intensity modulator;

[0009] Furthermore, a beam splitter BS is provided between the narrow linewidth laser and the intensity modulator in step 1, and the beam splitter BS splits the laser into two light beams, one light beam is used as the input light signal of the intensity modulator, and the other light beam is fed back to the laser for frequency compensation through the PDH frequency locking module; the narrow linewidth laser is a 1550nm narrow linewidth laser;

[0010] The initial quantum state preparation process directly measures the initial quantum state (coherent state). The probability of a photon being detected at any time within the coherent time is the same, that is, it appears as a completely uniform random distribution in the time domain, and this random distribution in the time domain has been proven to be incoherent with the frequency domain basis.

[0011] In the PDH frequency locking module, the phase of the reflected light signal from the ultra-stable cavity is demodulated to obtain the frequency detuning information of the reflected light, generate an error signal, and feed it back to the laser for frequency compensation, thereby locking the laser on the ultra-stable cavity and improving the coherence of the light source.

[0012] Step 2, quantum state manipulation; the time-varying signal is mapped to the coherent state through the intensity modulator, the probability of the modulated coherent state photon being detected is directly proportional to the signal, and the microwave signal to be measured modulates the coherent state photon through the electro-optical crystal; the bias voltage is adjusted through the DC bias controller, and a half-wave voltage is input to the electro-optical crystal to ensure that the electro-optical crystal works at the optimal linear working point;

[0013] The intensity modulator maps the time-varying signal to a coherent state, ensuring that the probability distribution of its measured collapse is proportional to the signal amplitude while maintaining the sparsity of the probability distribution function in the transform domain;

[0014] Step 3, quantum state detection: the light output by the electro-optical crystal enters the adjustable optical attenuator, and the light intensity is attenuated to the single-photon level by the adjustable optical attenuator, and then enters the single-photon detector, which detects the single-photon signal and outputs a pulse signal to the time interval analyzer, which measures the pulse signal and records the time sequence of the arrival of the photons;

[0015] Further, the reference clock of the time interval analyzer in step 3 is a GPS locked quartz crystal;

[0016] The observation operator in the quantum state detection process does not directly determine the observation matrix. Only when the measurement occurs, the random collapse of the quantum state forms the observation matrix. The observation matrix is ​​jointly determined by the initial state, the signal, and the observation operator.

[0017] Step 4, signal reconstruction: the time series collected by the time interval analyzer is transmitted to the data processing unit for storage, and the original signal is accurately restored from the partial measurement values ​​through the signal recovery algorithm.

[0018] Compared with the prior art, the present invention has the following advantages:

[0019] (1) This paper proposes a new digital signal processing method based on quantum compressed sensing. It uses quantum physics processes to build a compressed sensing system to achieve compressed measurement of sparse signals in the transform domain. The real-time measurement bandwidth reaches 20 GHz and the data compression rate reaches 10 -6 , showing a significant quantum advantage;

[0020] (2) The present invention adopts the quantum compressed sensing theoretical model. When measuring a signal, a measurement matrix is ​​constructed based on the initial quantum state, the measured signal, and the observation operator. Therefore, quantum compressed sensing is an adaptive measurement method.

[0021] (3) The present invention uses an electro-optical crystal to modulate the coherent photon wave function of the measured signal, which greatly reduces the sampling rate requirements for information recovery and improves the real-time bandwidth. The sparse photon time series is detected by a low time jitter single photon detector and a time interval analyzer, and the recovery algorithm is executed to obtain the spectrum. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 It is a flow chart of a digital signal processing method based on quantum compressed sensing of the present invention;

[0023] Figure 2 It is a schematic diagram of a device of a digital signal processing method based on quantum compressed sensing of the present invention;

[0024] Figure 3 It is a schematic diagram of microwave signal recovery results of a digital signal processing method based on quantum compressed sensing of the present invention;

[0025] Figure 4 It is a schematic diagram of the spectrum measurement results of a multi-frequency component signal according to a digital signal processing method based on quantum compressed sensing of the present invention. DETAILED DESCRIPTION

[0026] In order to gain a deeper understanding of the present invention, we will provide a comprehensive and detailed description of the present invention. However, the present invention has multiple implementations and is not limited to the specific examples listed herein. The presentation of these examples is intended to deepen the comprehensive understanding of the disclosure of the present invention.

[0027] A digital signal processing method based on quantum compressed sensing, the flow chart is as follows Figure 1 The schematic diagram of the device is shown in Figure 2 As shown, the method comprises the following steps:

[0028] A digital signal processing method based on quantum compressed sensing, the method comprising the following steps:

[0029] Step 1, initial quantum state preparation: the narrow linewidth laser outputs laser light into the intensity modulator;

[0030] Furthermore, a beam splitter BS is provided between the narrow linewidth laser and the intensity modulator in step 1, and the beam splitter BS splits the laser into two light beams, one light beam is used as the input light signal of the intensity modulator, and the other light beam is fed back to the laser for frequency compensation through the PDH frequency locking module; the narrow linewidth laser is a 1550nm narrow linewidth laser;

[0031] The initial quantum state preparation process directly measures the initial quantum state (coherent state). The probability of a photon being detected at any time within the coherent time is the same, that is, it appears as a completely uniform random distribution in the time domain, and this random distribution in the time domain has been proven to be incoherent with the frequency domain basis.

[0032] In the PDH frequency locking module, the phase of the reflected light signal from the ultra-stable cavity is demodulated to obtain the frequency detuning information of the reflected light, generate an error signal, and feed it back to the laser for frequency compensation, thereby locking the laser on the ultra-stable cavity and improving the coherence of the light source.

[0033] Step 2, quantum state manipulation; the time-varying signal is mapped to the coherent state through the intensity modulator, the probability of the modulated coherent state photon being detected is directly proportional to the signal, and the microwave signal to be measured modulates the coherent state photon through the electro-optical crystal; the bias voltage is adjusted through the DC bias controller, and a half-wave voltage is input to the electro-optical crystal to ensure that the electro-optical crystal works at the optimal linear working point;

[0034] The intensity modulator maps the time-varying signal to a coherent state, ensuring that the probability distribution of its measured collapse is proportional to the signal amplitude while maintaining the sparsity of the probability distribution function in the transform domain;

[0035] Step 3, quantum state detection: the light output by the electro-optical crystal enters the adjustable optical attenuator, and the light intensity is attenuated to the single-photon level by the adjustable optical attenuator, and then enters the single-photon detector, which detects the single-photon signal and outputs a pulse signal to the time interval analyzer, which measures the pulse signal and records the time sequence of the arrival of the photons;

[0036] Further, the reference clock of the time interval analyzer in step 3 is a GPS locked quartz crystal;

[0037] The observation operator in the quantum state detection process does not directly determine the observation matrix. Only when the measurement occurs, the random collapse of the quantum state forms the observation matrix. The observation matrix is ​​jointly determined by the initial state, the signal, and the observation operator.

[0038] Step 4, signal reconstruction: the time series collected by the time interval analyzer is transmitted to the data processing unit for storage, and the original signal is accurately restored from the partial measurement values ​​through the signal recovery algorithm.

[0039] Figure 3 This is a schematic diagram of the microwave signal recovery results of a digital signal processing method based on quantum compressed sensing of the present invention. As shown in the figure, the black curve is the original signal, and the red, green and blue dotted lines are the recovered 5GHz, 10GHz and 20GHz sine wave signals, with fidelity of 98.4%, 88.0% and 59.7% respectively.

[0040] Figure 4 This is a schematic diagram of the spectrum measurement results of a multi-frequency component signal of a digital signal processing method based on quantum compressed sensing of the present invention. The measurement spectrum range is set to 4-5GHz, and the amplitudes of all frequency components are set to the same, as shown in the figure. Figure 4 (a), (b), and (c) represent the spectrum diagrams with frequency components of 5, 25, and 50, respectively. It can be found that with the increase of frequency components, the amplitude of the characteristic spectrum line decreases significantly, the relative fluctuation range of the amplitude increases, and the white noise base remains unchanged, that is, the spectrum signal-to-noise ratio gradually decreases;

[0041] Figure 4 (d) shows the changing trend of the characteristic spectral line amplitude and the white noise base amplitude when the frequency component changes from 1 to 100. It can be seen that when the frequency component reaches 100, the quantum compressed sensing system can still distinguish the signal and the background well. Therefore, the quantum compressed sensing system can be applied to the measurement task of multiple frequency components to a certain extent, but as the frequency component increases, the signal-to-noise ratio will be significantly reduced.

[0042] The contents not described in detail in the specification of the present invention belong to the prior art known to the professional and technical personnel in the field. Although the illustrative specific embodiments of the present invention are described above to facilitate the understanding of the present invention by the technical personnel in the field, it should be clear that the present invention is not limited to the scope of the specific embodiments. For the ordinary technical personnel in the field, as long as various changes are within the spirit and scope of the present invention defined and determined by the attached claims, these changes are obvious, and all inventions and creations using the concept of the present invention are protected.

Claims

1. A digital signal processing method based on quantum compressed sensing, characterized in that: The method comprises the following steps: Step 1, initial quantum state preparation: the narrow linewidth laser outputs laser light into the intensity modulator; Step 2, quantum state manipulation; the time-varying signal is mapped to the coherent state through the intensity modulator, the probability of the modulated coherent state photon being detected is directly proportional to the signal, and the microwave signal to be measured modulates the coherent state photon through the electro-optical crystal; the bias voltage is adjusted through the DC bias controller, and a half-wave voltage is input to the electro-optical crystal to ensure that the electro-optical crystal works at the optimal linear working point; Step 3, quantum state detection: the light output by the electro-optical crystal enters the adjustable optical attenuator, and the light intensity is attenuated to the single-photon level by the adjustable optical attenuator, and then enters the single-photon detector, which detects the single-photon signal and outputs a pulse signal to the time interval analyzer, which measures the pulse signal and records the time sequence of the arrival of the photons; Step 4, signal reconstruction: the time series collected by the time interval analyzer is transmitted to the data processing unit for storage and the original signal is accurately restored from the partial measurement values ​​through the signal recovery algorithm.

2. A digital signal processing method based on quantum compressed sensing according to claim 1, characterized in that: A beam splitter BS is also provided between the narrow linewidth laser and the intensity modulator in step 1. The beam splitter BS splits the laser into two beams, one of which is used as the input optical signal of the intensity modulator, and the other is fed back to the laser for frequency compensation through the PDH frequency locking module.

3. A digital signal processing method based on quantum compressed sensing according to claim 2, characterized in that: The reference clock for the time interval analyzer in step 3 is a GPS locked quartz crystal.

4. The digital signal processing method based on quantum compressed sensing according to claim 3 is characterized in that: The narrow linewidth laser is a 1550nm narrow linewidth laser.

5. The digital signal processing method based on quantum compressed sensing according to claim 4, characterized in that: In the PDH frequency locking module, the ultra-stable cavity reflected light signal is phase demodulated to obtain the frequency detuning information of the reflected light, generate an error signal, and feed it back to the laser for frequency compensation, so as to lock the laser on the ultra-stable cavity and improve the coherence of the light source.

6. The digital signal processing method based on quantum compressed sensing according to claim 5, characterized in that: The intensity modulator maps the time-varying signal to the initial quantum state, ensuring that the probability distribution of its measured collapse is proportional to the signal amplitude, while maintaining the sparsity of the probability distribution function in the transform domain.

Images